The long-term objective of this research is to understand visual signaling by the primate retina, the most relevant model for human vision, and to exploit this knowledge in the treatment of blindness. The goal of the proposed work is to determine how diverse retinal ganglion cell (RGC) types in the retina signal visual information to the brain, and the mechanisms they use to extract this information from the visual environment. We have identified two RGC types, ON and OFF SM cells, that show high promise for understanding how cell type diversity contributes to visual signaling in primates. We have shown that these cell types exhibit unusual spatial properties and high stimulus selectivity, reminiscent of the unique visual computations carried out by RGCs in other species. Based on these findings, our specific aims are to 1) Determine the structure and anatomical substrate of the unique spatial computations performed by SM cells; 2) Determine the cellular and synaptic mechanisms that underlie this spatial signaling as well as the sparse firing of SM cells; and 3) Determine if the unique spatial structure and sparse firing in SM cells underlie specialized computations that may subserve specific visual functions. To approach these problems, we will collaboratively combine large-scale multi-electrode recordings with intracellular recordings and high resolution anatomical identification of SM cells in isolated macaque retina, a unique and powerful collection of methods ideally suited to the task. At the conclusion of this work we expect to understand the visual computations performed by SM cells and the mechanisms that mediate these computations.